Repository 'differential_count_models'
hg clone https://testtoolshed.g2.bx.psu.edu/repos/fubar/differential_count_models

Changeset 124:731315bd6e48 (2014-11-25)
Previous changeset 123:51f998262ada (2014-11-25) Next changeset 125:7e0b4a60da89 (2014-11-27)
Commit message:
Uploaded
modified:
rgedgeRpaired_nocamera.xml
b
diff -r 51f998262ada -r 731315bd6e48 rgedgeRpaired_nocamera.xml
--- a/rgedgeRpaired_nocamera.xml Tue Nov 25 05:36:47 2014 -0500
+++ b/rgedgeRpaired_nocamera.xml Tue Nov 25 05:50:07 2014 -0500
[
b'@@ -8,7 +8,7 @@\n   </requirements>\n   \n   <command interpreter="python">\n-     rgToolFactory.py --script_path "$runme" --interpreter "Rscript" --tool_name "DifferentialCounts" \n+     rgToolFactory.py --script_path "$runme" --interpreter "Rscript" --tool_name "Differential_Counts" \n     --output_dir "$html_file.files_path" --output_html "$html_file" --make_HTML "yes"\n   </command>\n   <inputs>\n@@ -22,12 +22,12 @@\n     </param>\n     <param name="treatment_name" type="text" value="Treatment" size="50" label="Treatment Name"/>\n     <param name="Treat_cols" label="Select columns containing treatment." type="data_column" data_ref="input1" numerical="True" \n-         multiple="true" use_header_names="true" size="120" display="checkboxes">\n+         multiple="true" use_header_names="true" size="120" display="checkboxes"  force_select="True">\n         <validator type="no_options" message="Please select at least one column."/>\n     </param>\n     <param name="control_name" type="text" value="Control" size="50" label="Control Name"/>\n     <param name="Control_cols" label="Select columns containing control." type="data_column" data_ref="input1" numerical="True" \n-         multiple="true" use_header_names="true" size="120" display="checkboxes" optional="true">\n+         multiple="true" use_header_names="true" size="120" display="checkboxes"  force_select="True">\n     </param>\n     <param name="subjectids" type="text" optional="true" size="120" value = ""\n        label="IF SUBJECTS NOT ALL INDEPENDENT! Enter comma separated strings to indicate sample labels for (eg) pairing - must be one for every column in input"\n@@ -85,47 +85,6 @@\n       <option value="F" selected="true">Do not run VOOM</option>\n       <option value="T">Run VOOM</option>\n      </param>\n-    <!--\n-    <conditional name="camera">\n-        <param name="doCamera" type="select" label="Run the edgeR implementation of Camera GSEA for up/down gene sets" \n-            help="If yes, you can choose a set of genesets to test and/or supply a gmt format geneset collection from your history">\n-        <option value="F" selected="true">Do not run GSEA tests with the Camera algorithm</option>\n-        <option value="T">Run GSEA tests with the Camera algorithm</option>\n-        </param>\n-         <when value="T">\n-         <conditional name="gmtSource">\n-          <param name="refgmtSource" type="select" \n-             label="Use a gene set (.gmt) from your history and/or use a built-in (MSigDB etc) gene set">\n-            <option value="indexed" selected="true">Use a built-in gene set</option>\n-            <option value="history">Use a gene set from my history</option>\n-            <option value="both">Add a gene set from my history to a built in gene set</option>\n-          </param>\n-          <when value="indexed">\n-            <param name="builtinGMT" type="select" label="Select a gene set matrix (.gmt) file to use for the analysis">\n-              <options from_data_table="gseaGMT_3.1">\n-                <filter type="sort_by" column="2" />\n-                <validator type="no_options" message="No GMT v3.1 files are available - please install them"/>\n-              </options>\n-            </param>\n-          </when>\n-          <when value="history">\n-            <param name="ownGMT" type="data" format="gmt" label="Select a Gene Set from your history" />\n-          </when>\n-          <when value="both">\n-            <param name="ownGMT" type="data" format="gseagmt" label="Select a Gene Set from your history" />\n-            <param name="builtinGMT" type="select" label="Select a gene set matrix (.gmt) file to use for the analysis">\n-              <options from_data_table="gseaGMT_4">\n-                <filter type="sort_by" column="2" />\n-                <validator type="no_options" message="No GMT v4 files are available - please fix tool_data_table and loc files"/>\n-              </options>\n-            </param>\n-           </when>\n-         </conditional>\n-         </when>\n-         <whe'..b'=\'_\'))},T)\n    }\n   if (doDESeq2 == T) {  DESeqcounts = run_DESeq2(workCM,pdata,subjects,group,out_DESeq2,mt,DESeq_fitType) }\n   if (doVoom == T) { voomcounts = run_Voom(workCM,pdata,subjects,group,mydesign,mt,out_Voom) }\n@@ -823,25 +782,23 @@\n     \n     if (nrow(counts.dataframe > 1)) {\n       counts.venn = vennCounts(counts.dataframe)\n-      vennf = paste("Venn",mt,"significant_genes_overlap.pdf",sep="_") \n+      vennf = paste("Differential_venn",mt,"significant_genes_overlap.pdf",sep="_") \n       pdf(vennf)\n       vennDiagram(counts.venn,main=vennmain,col="maroon")\n       dev.off()\n     }\n   } #### doDESeq2 or doVoom\n-  \n+sink()\n }\n #### Done\n ]]>\n-\n-###sink(stdout(),append=T,type="message")\n builtin_gmt = ""\n history_gmt = ""\n history_gmt_name = ""\n out_edgeR = F\n out_DESeq2 = F\n out_Voom = "$out_VOOM"\n-edgeR_robust_meth = "ordinary" # control robust deviance options\n+edgeR_robust_meth = "ordinary" \n doDESeq2 = $DESeq2.doDESeq2\n doVoom = $doVoom\n doCamera = F\n@@ -883,7 +840,7 @@\n fdrtype = "$fdrtype"\n fdrthresh = $fdrthresh\n useNDF = $useNDF\n-fQ = $fQ\n+fQ = $fQ # non-differential centile cutoff\n myTitle = "$title"\n sids = strsplit("$subjectids",\',\')\n subjects = unlist(sids)\n@@ -895,12 +852,13 @@\n cat(\'; CCols=\')\n cat(CCols)\n cat(\'\\n\')\n+<![CDATA[\n useCols = c(TCols,CCols)\n if (file.exists(Out_Dir) == F) dir.create(Out_Dir)\n Count_Matrix = read.table(Input,header=T,row.names=1,sep=\'\\t\') \n snames = colnames(Count_Matrix)\n nsamples = length(snames)\n-if (nsubj &gt; 0 &amp; nsubj != nsamples) {\n+if (nsubj >  0 & nsubj != nsamples) {\n options("show.error.messages"=T)\n mess = paste(\'Fatal error: Supplied subject id list\',paste(subjects,collapse=\',\'),\n    \'has length\',nsubj,\'but there are\',nsamples,\'samples\',paste(snames,collapse=\',\'))\n@@ -908,21 +866,23 @@\n quit(save="no",status=4)\n }\n if (length(subjects) != 0) {subjects = subjects[useCols]}\n-Count_Matrix = Count_Matrix[,useCols]\n+Count_Matrix = Count_Matrix[,useCols] ### reorder columns\n rn = rownames(Count_Matrix)\n islib = rn %in% c(\'librarySize\',\'NotInBedRegions\')\n LibSizes = Count_Matrix[subset(rn,islib),][1] # take first\n Count_Matrix = Count_Matrix[subset(rn,! islib),]\n group = c(rep(TreatmentName,length(TCols)), rep(ControlName,length(CCols)) )             \n group = factor(group, levels=c(ControlName,TreatmentName))\n-colnames(Count_Matrix) = paste(group,colnames(Count_Matrix),sep="_")                  \n+colnames(Count_Matrix) = paste(group,colnames(Count_Matrix),sep="_")        \n results = edgeIt(Count_Matrix=Count_Matrix,group=group, out_edgeR=out_edgeR, out_Voom=out_Voom, out_DESeq2=out_DESeq2,\n                  fdrtype=\'BH\',mydesign=NULL,priordf=edgeR_priordf,fdrthresh=fdrthresh,outputdir=\'.\',\n-                 myTitle=myTitle,useNDF=F,libSize=c(),filterquantile=fQ,subjects=subjects,\n+                 myTitle=myTitle,useNDF=F,libSize=c(),filterquantile=fQ,subjects=subjects,TreatmentName=TreatmentName,ControlName=ControlName,\n                  doDESeq2=doDESeq2,doVoom=doVoom,doCamera=doCamera,doedgeR=doedgeR,org=org,\n                  histgmt=history_gmt,bigmt=builtin_gmt,DESeq_fitType=DESeq_fitType,robust_meth=edgeR_robust_meth)\n sessionInfo()\n \n+sink()\n+]]>\n </configfile>\n </configfiles>\n <help>\n@@ -936,9 +896,11 @@\n \n Requires a count matrix as a tabular file. These are best made using the companion HTSeq_ based counter Galaxy wrapper\n and your fave gene model to generate inputs. Each row is a genomic feature (gene or exon eg) and each column the \n-non-negative integer count of reads from one sample overlapping the feature. \n+non-negative integer count of reads from one sample overlapping the feature.\n+\n The matrix must have a header row uniquely identifying the source samples, and unique row names in \n the first column. Typically the row names are gene symbols or probe ids for downstream use in GSEA and other methods.\n+They must be unique and R names or they will be mangled - please read the fine R docs for the rules on identifiers. \n \n **Specifying comparisons**\n \n'